1) Our state legislature is considering en equal rights amendment to the state constitution. Historically, it has 75% support among voters

• If we propose a poll of its popularity with the typical margin of error of 3% how many voters should we plan to interview to be 95 percent certain of the result, based on a "best guess" And what "safe" number of interviews would we propose

2) The Senate is thinking about average pay "gaps" Our research indicates that by profession, nearly all women earn the range from $12000 to 90000. The senate wants the Margin of Error to be no more than $750.

• How many subjects should be tested if the senate is looking to announce its finding with 99% confidence?

3) A survey on the gender pay gap showed men (52%) to be less in favor than women (55%) of passing a bill to penalize companies with pay gaps. With 3500 subjects in the survey, the result was Statistically Significant at the 5% level. The women's right groups thinks it should change its message to target male voters. What should we tell them about Statistically and Practically Significant result?